AivexaNewsSearch
AI news for builders and product teamsChecked every hour

After the midterms, things are looking rocky for Elon Musk and data centers

Collected Nov 7, 2026

Democrats won this week's midterm elections, and on the latest episode of TechCrunch's Equity podcast, hosts Kirsten Korosec, Sean O'Kane and Anthony Ha worked through what the result means for the tech industry. Their focus was data centers: in dozens of local ballot measures voters rejected the facilities, and in other races they elected politicians who campaigned on restricting them. Texas governor-elect Gina Hinojosa, who defeated Greg Abbott, was the example they kept returning to, having spoken openly about limits as a campaign promise she intends to pursue in office.

O'Kane framed the result around Elon Musk, described on the show as one of the biggest data center builders and the largest donor to the Republican Party. His read was that Musk got essentially nothing from donating half a billion dollars, possibly more once the final count is in, and now has all his businesses headquartered in a state that has turned about as blue as it could, for the first time in a very long time if ever. A correction issued during the conversation is worth noting for anyone citing the episode: early in the discussion a host said the ballot measures would have paved the way for faster data center construction and that almost all were shot down. That was walked back, with the clarification that many of the measures would instead have imposed restrictions or moratoriums, and almost all of those passed.

O'Kane said the consensus view in the tech world is that a lot of roadblocks now lie ahead for data centers, particularly in places like Texas. He also suggested the industry may be over-focused on the issue. Ha pushed back partly: the ballot measures were numerous but not on the ballot everywhere, and different races had different dynamics. Still, he saw an undercurrent that reaches across the country and is bipartisan in many ways, with Democrats appearing more effective at harnessing the energy. He argued anti-AI and especially anti-data center sentiment was a big part of why Tuesday went the way it did, and said the next two years, potentially longer, will be harder for tech companies generally and for hyperscalers and data center companies in particular. He added that those companies have to be worried about whether the issue persists into 2028 and beyond.

Korosec was more measured on causation. She did not think data centers brought everyone to the polls or produced what she called the quote-unquote blue wave, but said the issue has become increasingly grassroots, partly because of how hyperscalers and companies have gone about getting projects approved: very secretively, often requiring NDAs of elected or appointed officials, and not sharing information about the scale of a project in environmentally sensitive areas, including the Southwest, where water scarcity is a serious concern. Republicans, independents and Democrats alike saw that as wrong and were motivated by it, she said, without it being a singular issue.

Her most concrete observation concerned OpenAI. Amid claims that AI companies could not operate under any other circumstances, she said OpenAI has now released what it calls a compute campus policy blueprint, under which it will no longer require NDAs and will share its plans and supposedly operate with less secrecy. Her takeaway: when regulators supply certainty in the marketplace, or at least signals in that direction, corporations respond; without that, they will not necessarily do the hoped-for thing. Ha noted the blueprint is also a rebranding exercise, since data centers have become such a toxic phrase that they are now compute campuses, and wondered whether the label would take. O'Kane's answer was blunt: no, it will not change anything and will only cost the company effort, from a company that likes to say it is no longer focused on side quests. Korosec joked that OpenAI should create a cutesy compute campus avatar and present that instead of a power- and water-sucking data center; Ha suggested the companies build their own offices there or work next door if the campuses are so lovely.

Ha raised a second pressure point: the election may be the second part of a one-two punch after a shift in how Wall Street reacts to hyperscalers and AI companies, with investors unenthusiastic about spending so heavily on new data centers and compute. That creates a financial incentive not to build. The squeeze, as he described it, is that regulators will probably make construction harder and the market probably will not reward it, yet reaching the scale these companies want probably requires building data centers somewhere.

Why it matters: Hyperscalers and the developers who depend on their capacity face a tighter path to new sites, with local ballots, state officials and, by the hosts' reading, skittish investors all pushing the same direction. The OpenAI blueprint suggests permitting posture and transparency are becoming competitive levers rather than messaging, which is an inference from the discussion rather than a stated finding. Teams planning large compute builds should expect longer approval cycles and more public scrutiny, especially around water use in arid regions.

Read at TechCrunch · AI

Based on reporting from the original publisher. Visit the source for full context and later updates.

Publisher excerpt

On Equity, we unpacked what the midterms mean for the tech industry.